Leadership and Companies
Google Brings Background Voice Tasks and Visual Context to Gemini
Google introduced two Gemini models designed for more natural voice conversations and complex tasks requiring background tool execution.
What happened
Google launched Gemini 3.8 Live for fast, fluid dialogue and Gemini 3.8 Live Extended Thinking for complex, multi-step tasks. Both models support near-real-time visual context, background tool and API calls, and automatic switching across 97 languages. They are available through the Gemini API, Google AI Studio, Gemini, Workspace, Search, and selected enterprise previews.
Why it matters
Developers and enterprises can build voice agents that continue conversations while completing tasks in the background, with the extended model aimed at higher-complexity workflows.
Source: Google Blog